Weak spark -really baffled

[Riick]

Hi, hoping someone on this forum can help. Been working on this for 2+ months and it's driving me crazy. I'm a car repair noob so I might have missed something simple. Any thoughts appreciated!

The car won't start because of a bad spark (sometimes weak, sometimes absent, occassionally pops out one nice spark).

First I thought it was the coil because of scorch marks, but a brand new coil didn't help.
Then I saw oil on a sensor in the distributor housing so I put in a used distributor (and a used ECU); worked great for one day.
Then I noticed that the engine block was getting 1 volt, which I traced to G1. Cleaned that off (and also G101 and G3) and it worked great for one day.

No oil on distributor sensors. Getting 12V at coil + and coil -. Getting 9.5V at signal wire. Getting 0.1V at engine block. Test light flashes when cranking (although it only flicks off for just an instant each cycle- is that normal?) I 'm testing the spark at the coil with spark tester gapped 1/8 inch.

Both days it worked I had made some of the ground wires wet while cleaning, but the negligible engine voltage even after things are dry makes me think that may not be it. I might try jumping wires to battery negative (from block, ICM heat sink, each of 4 ECU ground wires) just to confirm. What else can I try here?

 

[RonJ]

Have an auto parts store test the ICM.

Resistance test the coil.

What's the resistance between G101 and G1?

Does the fuel pump always prime?

Any CEL codes?

 

[Riick]

No codes.

Fuel pump primes every time I'm listening for it.

Coil has 16,000 ohms from coil negative to secondary. I can't measure fractional ohms with my multimeter but coil positive to coil negative is very close to zero. (It seems unlikely that the brand new coil plus my two new used ones would have all failed there although I suppose it's possible)

I'll check the G101 to G1 resistance tomorrow.

It will be a while until I can get to the auto parts store... is it still possible the ICM is bad even though the test light flashes at the output? Is there some way I can test this myself?

 

[mymmeryloss]

Do u have good chassis to engine grounds?

Battery charged?

 

[TigBitties]

check your trans ground too

 

[RonJ]

No codes.

Does the CEL work?

Coil has 16,000 ohms from coil negative to secondary. I can't measure fractional ohms with my multimeter but coil positive to coil negative is very close to zero. (It seems unlikely that the brand new coil plus my two new used ones would have all failed there although I suppose it's possible)

It's important to rule out the coil as it's the most common cause of weak spark. What is the primary coil resistance and what is the resistance read when you touch the two meter probes together?

It will be a while until I can get to the auto parts store... is it still possible the ICM is bad even though the test light flashes at the output? Is there some way I can test this myself?

Yes, the ICM could still be bad. Here are some tests. Read the entire post and test for bleed:

http://www.tegger.com/hondafaq/igniter-operation/off-car-testing.html

 

[mymmeryloss]

Ron, ur the man....

 

[Riick]

Info: I tried grounding the engine block directly with jumpers, and also the wires at G101 (hard to do), and also the back of the ICM. This gave me some nice bright sparks, but it was inconsistent; sometimes when I cranked it was just a few bright sparks, other times nearly perfect, other times no sparks of any kind. I couldn't find much relation to how I had arranged the jumper cables.

What's the resistance between G101 and G1?

between 0 and 20 ohm. My analog multimeter is not precise for low resistances.

Battery charged?

Yes. 13V.

Do u have good chassis to engine grounds?

It's good. No voltage drop between engine block and chasis when cranking.

check your trans ground too

Confused... Isn't the transmission ground the same as the engine ground? The only other ground to chasis goes to the valve cover and I thought not much used that?

Does the CEL work?

Yes. (Gives injector codes when I unplug the injectors.)

What is the primary coil resistance and what is the resistance read when you touch the two meter probes together?

Between 0 and 20 ohm. My analog multimeter is not precise for low resistances, but it looks the same in both cases.

Sounds like I need to get me a better multimeter! That may take a while. Meanwhile I'll see if I can test the ICM.

 

[RonJ]

You can buy an adequate digital multimeter for $3-$4 at Harbor freight Tools. You need one for your car.

 

[Riick]

Found some oxidation on igniter contacts. I won't know if this was the problem until I recharge my battery (engine turning slow at 11.95 V). I'll try to buy a charger tomorrow.

Igniter passed the do-it-yourself test but only after oxidation sanded off (thanks for the link Ron). Multimeter in place of coil gets battery voltage for grounded signal line, 0V for floating signal line, 0V for battery voltage signal line.

What's the resistance between G101 and G1?

0.1 ohms

What is the primary coil resistance

0.8 ohms

 

[RonJ]

0.8 ohms

What's the internal resistance of the multimeter (see quote below)? Or did you already subtract it from the primary coil resistance?

what is the resistance read when you touch the two meter probes together?

 

[Riick]

What's the internal resistance of the multimeter (see quote below)? Or did you already subtract it from the primary coil resistance?

My new multimeter does something funny with that. It consistently reads 0.0 ohms when probes are touching the same piece of clean bare metal, but it often reads 0.2 ohms when probes are touching each other, unless I hold them away from the multimeter. But clean bare metal is 0.0 ohms every time.

 

[RonJ]

At this point, I recommend that you re-assemble the distributor and then test spark at the spark plugs. Use the method described at the spark link in my signature. Are your spark results any different?

 

[Avant Uprising]

At this point, I recommend that you re-assemble the distributor and then test spark at the spark plugs. Use the method described at the spark link in my signature. Are your spark results any different?

^ this and how are your grounds at the thermostat housing too?

 

[Riick]

At this point, I recommend that you re-assemble the distributor and then test spark at the spark plugs. Use the method described at the spark link in my signature. Are your spark results any different?

Not any better. (Spark weak and sometimes absent. Pops out one nice spark every once in a while.)

I also checked spark at one of the four distributor terminals (where the high tension wires connect to the distributor)- same poor results.

The spark at the coil is pretty good now (jumps 4mm with no effort).

Is it possible that my cap or my rotor failed even though they are less than one year old? Even though I don't see any obvious "carbon tracks"?

^ this and how are your grounds at the thermostat housing too?

I think the thermostat ground (G101) is fine. I sanded off all the corrosion last week and have been getting 0.1 ohms between the negative battery terminal and the each of the four ground wires at the computer plugs.

 

[RonJ]

I also checked spark at one of the four distributor terminals (where the high tension wires connect to the distributor)- same poor results.

The spark at the coil is pretty good now (jumps 4mm with no effort).

What color is spark at the coil? If it's bright white and consistent, then the coil spring, cap, and/or plug wires may be bad.

 

[Riick]

What color is spark at the coil?

It's white and always present now, but it gets a little dimmer sometimes (I think only after cranking a few seconds, but not sure.) No trouble jumping 4mm gap even when dimmer.

If it's bright white and consistent, then the coil spring, cap, and/or plug wires may be bad.

Could the rotor be leaking some voltage to the shaft?

Can I diagnose further? Or do I just buy and try at this point?

 

[Riick]

Problem solved! (Actually it was two problems)

Cleaning the igniter contacts gave me a bright and consistent spark at the coil, but I still had barely any spark at the "towers" on the cap.

Replacing the rotor allowed the spark to make it through the cap and out to the end of the wires. My old aftermarket rotor was less than a year old and looked clean and perfect, but I suspect there was arc-through to the distributor shaft from some factory defect in the plastic. I noticed my new honda rotor has fiberglass plates on each side of the metal in order to prevent this problem.

Thanks to everyone for their help! And Ron that igniter link you sent was a real breakthrough- you are a wealth of helpful information.